ISIS Experimental Report RB Number: 820064 Rutherford Appleton Laboratory Date of Report: 14 Oct. 2008 Title of Experiment: The effect of pressure on the structure of Local Contact: M. G. Tucker meridianiite (MgSO 4·11 H2O) Principal Proposer: A. D. Fortes Instrument: PEARL Affiliation: University College London Experimental Team: A. D. Fortes, I. G. Wood, H. E. A.Brand Date of 11-14/09/2008 Experiment:

Introduction: The title substance is a relatively poorly tons (~10 kbar from the Pb pressure marker), the diffraction characterised material. However, it has been the subject of pattern of meridianiite changed substantially from that observed greater scrutiny in the past few years since it has been suggested at 27 tons (Fig. 1). We could not ascertain positively the that it may be an important on [1], and also that it identity of the new phase or phase mixture; several peaks could may be a major rock-forming mineral in the outer solar system be attributed to ice VI, suggesting that meridianiite has [2]. It was recently identified in terrestrial frozen brine ponds undergone a dehydration reaction to ice plus a lower hydrate [3], and named meridianiite (for on Mars). (possibly, but not necessarily, ). Warming from 240 K Peterson and Wang [4] solved the structure from single-crystal to 250 K resulted in a further change in the diffraction pattern. X-ray diffraction data. We have carried out a range of No further changes were seen until the temperature was experimental and computational studies of meridianiite, increased to 290 K, where a further apparent phase change was including a neutron powder diffraction study over the detected. As there is no obvious agreement between the temperature range 4.2 – 250 K on HRPD [5] and a single-crystal observed diffraction data and that of high-pressure epsomite study of the protonated phase on SXD (see RB820008). powder patterns observed by us previously on PEARL/HiPr it is However, we have no information concerning the phase possible that these patterns represent new polymorphs of behaviour or elasticity of meridianiite under the conditions meridianiite, or some phase with a hydration number between appropriate to the interiors of icy moons (0 – 5 GPa, 100 – 300 that of meridianiite (11) and epsomite (7). K). The aim of this work was to carry out an initial exploration in the relevant P-T regime using the Paris-Edinburgh cell, making observations that will form the foundations for future detailed study, and for comparison with the phase behaviour of other hydrates in the MgSO 4 – H2O system (see RB520039). Sample preparation: Meridianiite was formed by rapid quenching with L-N2 of a stoichiometric (supersaturated) solution of MgSO 4 in D 2O. Slugs of fine polycrystalline material formed in this way were broken up and ground to a powder in the UCL Sciences cold rooms. Since the ISIS cold-rooms could not be reduced in temperature below +6° C (meridianiite decomposes at +2° C), we spent 9 hours attempting to load the specimen at room temperature, using dry-ice to cool the anvils of the PE press (with an argon-filled glove bag to avoid icing). Despite considerable effort, this method failed (the sample decomposed), and we instead brought a pair of anvils back to UCL and loaded the specimen (at -15° C) in the Earth Sciences cold room. Due to uncertainty regarding the behaviour of the specimen in MeOD/EtOD, dry powder was loaded into the gaskets, and this resulted in a relatively poor tranmission of pressure under modest loads; the sample did not begin to take up the load until it was increased to > 20 tons. Results: Despite the very low symmetry (space group P1 ), diffraction patterns obtained at 177 K under a load of 7 tons exhibiting strong Bragg reflections close to the instrumental resolution, suitable for unit-cell refinement. Increasing the load Fig. 1. Changes in the diffraction pattern of meridianiite at 200 K resulted in considerable line broadening, and so the indicative of a phase change on loading from 27 to 31 tons at specimen was warmed to 240 K, whereupon the Bragg peaks 240 K, and subsequent warming up to room temperature under a became sharp once more. We determined that with load load of 31 tons. increments of 3 tons we could follow the evolution of the unit- cell parameters without difficulty. The bulk modulus found in References: this initial study by fitting an equation of state to the refined [1] Chou & Seal (2007): J. Geophys. Res. Planets . 112 , article E11004. unit-cell parameters in the range 0 – 7 kbar is in excellent [2] Kargel (1991): Icarus 94 (2), 369-390. [3] Peterson et al ., (2007): American Min . 92 (10), 1756-1759. agreement with that found in our gas-cell study on HRPD [4] Peterson &Wang (2006): Geology 34 (11), 957-960. (RB810006) and with ab initio calculations. Under a load of 31 [5] Fortes et al ., (2008): Phys. Chem. Min . 35 (4),207-221.